The present invention relates to robot control methods of the playback type and to apparatuses which employs the methods. More particularly, the invention relates to a robot control method and apparatus in which position data and data regarding velocity are taught at a plurality of points on an operation path of a work program, so that the taught data can be played back to drive a servo system and thereby repeat the work program.
In a conventional robot control method of the playback type, the taught data for the robot comprises position data for the teaching points and data regarding the velocity from a previous teaching point to the present teaching point. The taught data regarding the velocity is in the form of the absolute time for movement in physical space, such as a value determining the time for moving between two teaching points, for example, a mm/s or a seconds.
In the case where the data regarding the velocity is taught in terms of a value determining the time for moving between two teaching points, however, the following problems arise.
The data regarding the velocity determines only how much time is taken in moving between two points. If operation is connected at a standstill at a certain point, such as during performance of function processing without moving, inputting and outputting of signals with respect to peripheral devices, and so on, time is needed for interpretation of function information and input and output processing of signals. That time depends upon the processing speed of a control device. Further, in the servo system of the robot, there always exists a dynamic delay between a command position and the present position, so even the taught time for moving between the two points is not known if a playback is not executed. From these points of view, the cycle time of the work program is not accurately known if it is not actually measured at playback. In addition, if an attempt is made to determine the cycle time by actual measurement, the cycle time may vary depending upon the ambient temperature and fluctuations in the power source voltage. Accordingly, it is extremely difficult to control the cycle time.
Moreover, if a previously taught time for moving between the two points is altered, it is necessary to alter the velocity data per se. In the case where an attempt is made to shorten the cycle time of the work program, all the taught velocity data must be altered.